More than 30 million American smokers visit a primary care clinician each year. As a result, health care organizations, including health care delivery systems and front-line practice teams are uniquely positioned to help smokers successfully quit by facilitating their use of evidence-based tobacco cessation treatments. The 2000 Public Health Service Clinical Practice Guideline for Treating Tobacco Use and Dependence includes an evidence-based 5-A (Ask,Advise, Assess, Assist, Arrange) algorithm for providers to address tobacco dependence and help smokers to quit using proven treatments (Fiore et al., 2000). However, the 2005 National Health Interview Survey (NHIS) indicates that, among the 49% of smokers who attempt cessation each year, fewer than 5% used proven behavioral treatment, less than 1/3 used pharmacological aids, and fewer than 4% used the gold standard of combined behavioral and pharmacological interventions (Curry et al., 2007). One hypothesis is that the complexity and time demands of the 5-A algorithm may discourage providers from addressing tobacco use during health care visits and that a simpler two-step (Ask and Act) approach may increase provider advice and assistance and, consequently, rates of evidence-based treatment use among smokers (Schroeder, 2005). This study directly tests this hypothesis in community-based primary care practice settings. The specific aims are: (1) To determine the effectiveness of electronic medical record (EMR) prompting for the 5-A and 2-A clinical practice algorithms. Using a three-arm group randomized trial, the two algorithms will be compared to each other and to a no-EMR prompt control condition. Primary care practice teams comprise the unit of randomization and analysis. Outcome measures are rates of referral of smokers to behavioral interventions and pharmacotherapy treatments determined by longitudinal data obtained from the EMR, provider self-reports, patient self-reports, and the Wisconsin Tobacco Quit Line. (2 To test whether certain practice team characteristics moderate or mediate the effectiveness of the clinical algorithms. (3) To estimate patient-level outcomes associated with EMR prompting, including use of behavioral or pharmacological interventions, smoking cessation and sustained abstinence. (4) To estimate the cost-effectiveness of the 5-A and 2-A clinical algorithms with reagrd to cost-per-treatment utilization.